Indomethacin

What it is

Indomethacin is a prescription non-steroidal anti-inflammatory drug. It is taken by mouth as capsules or an oral suspension, given as a suppository, or given intravenously to newborn babies. It blocks the cyclooxygenase enzymes COX-1 and COX-2, which reduces the body's production of prostaglandins. It is one of the oldest NSAIDs still in wide use and is generally regarded in the comparative-safety literature as one of the more toxic members of the class.

What the research says

For the conditions it is approved for, the evidence is strongest for short-term symptom relief rather than for changing the course of disease. Systematic reviews of NSAIDs as a class show clear short-term pain and function benefit in axial spondyloarthritis; the placebo-controlled evidence in acute gout rests on a single very small trial and is rated low certainty. In preterm babies, indomethacin reliably closes or prevents a patent ductus arteriosus, but the largest randomised trial found no improvement in survival free of neurosensory impairment at 18 months. Against those benefits sit a boxed warning for heart attack, stroke, and gastrointestinal bleeding, and observational meta-analyses that place indomethacin among the higher-risk NSAIDs for both.

Evidence grade: Well established.

How it works

Drug class: Non-selective non-steroidal anti-inflammatory drug (NSAID) of the indole acetic acid class; inhibitor of cyclooxygenase COX-1 and COX-2

Indomethacin blocks the two cyclooxygenase enzymes, COX-1 and COX-2, so the body makes fewer prostaglandins. Prostaglandins make nerve endings more sensitive to pain and drive inflammation, so blocking them reduces pain, swelling and fever. The same enzymes protect the stomach lining and help maintain blood flow through the kidneys, which is why blocking them causes the drug's main harms. (Source 1)

Boxed warning

WARNING: RISK OF SERIOUS CARDIOVASCULAR AND GASTROINTESTINAL EVENTS

(Source 2)

What it is used for

• A Cochrane review of NSAIDs as a class found high-quality evidence of better pain relief than placebo at six weeks, with pain 16.5 points lower on a 100 mm scale and a number needed to treat of 4. The review pooled NSAIDs together rather than reporting indomethacin separately, and no trial in the meta-analyses included people with non-radiographic axial spondyloarthritis. Evidence: established. (Source 3)
• The 2021 Cochrane review found only one placebo-controlled trial of an NSAID in acute gout, with 30 participants, and rated the evidence low certainty. Eleven of 15 on the NSAID versus 4 of 15 on placebo had a 50% pain reduction at 24 hours, but there was little to no effect on swelling by day four, and adverse events showed no difference either way. The review pools NSAIDs as a class and reports nothing for indomethacin on its own. The label's own dosing text describes pain relief within 2 to 4 hours. Evidence: limited. (Source 4)
• This is an approved indication on the 2026 US label, but we did not retrieve a systematic review of indomethacin specifically in rheumatoid arthritis. The label's own adverse-reaction table draws on 33 double-blind controlled trials reported in the literature, so a trial base exists, but we did not read those trials. Evidence: limited. (Source 5)
• An approved indication on the 2026 label. We did not retrieve an osteoarthritis-specific systematic review for indomethacin; the class-level randomised evidence we did read is about harms rather than osteoarthritis benefit. Evidence: limited. (Source 5)
• An approved indication for a short course. We found no trial or systematic review for this use in the literature we searched, so the evidence behind it is unknown to us. Evidence: unknown. (Source 5)
• Indomethacin clearly reduces patent ductus arteriosus. A 2022 meta-analysis of randomised trials of prophylactic indomethacin found an odds ratio of 0.31, and the 1202-infant TIPP trial found 24% versus 50% in the placebo group. The ductal effect is not in doubt; the longer-term benefit is. Evidence: established. (Source 6)
• The TIPP trial, the largest randomised trial, found no improvement in survival without neurosensory impairment at 18 months: 47% versus 46% died or survived with impairment, odds ratio 1.1, P=0.61. A 2022 meta-analysis found no significant effect on bronchopulmonary dysplasia, pulmonary haemorrhage, intraventricular haemorrhage, necrotising enterocolitis, intestinal perforation, mortality or length of stay. Evidence: not-supported. (Source 7)
• Response to indomethacin is treated as the defining feature of these two headache disorders, but the published evidence is case series rather than controlled trials. In a tertiary-centre series, eleven of eighteen patients (61%) responded, with a median 13 days to meaningful response in hemicrania continua, most at 150 mg/day. Evidence: limited. (Source 8)
• A Cochrane review of 20 studies and 1509 women, with indomethacin used in 15 of them, concluded that no clear benefit for COX inhibitors was shown over placebo or any other tocolytic. There was no difference in birth within 48 hours of trial entry (RR 0.20, 95% CI 0.03 to 1.28), and the authors judged the studies generally low quality. Evidence: limited. (Source 9)

Interactions

• Alcohol (label): Drinking alcohol while taking indomethacin adds to the chance of bleeding or ulceration in the stomach or gut. The label lists alcohol use among the factors that increase gastrointestinal bleeding risk on NSAIDs, alongside corticosteroids, aspirin, anticoagulants, SSRIs, smoking and older age. The evidence recorded here is the label statement; the case-control study of alcohol with aspirin and ibuprofen is carried separately as a finding, because the interaction vocabulary has no value for a case-control study. (Source 10)
• Potassium-sparing diuretics, and by the same mechanism potassium supplements (label): Indomethacin itself can push blood potassium up. Combining it with anything else that raises potassium, such as a potassium-sparing diuretic, can add to that. The label addresses potassium-sparing diuretics specifically; it does not address potassium supplements, so that part is mechanism rather than measured evidence. (Source 11)
• Triamterene (clinical trial): Adding the potassium-sparing diuretic triamterene to ongoing indomethacin caused reversible acute kidney failure in two of four healthy volunteers, and the label says the two should not be given together. (Source 11)
• Warfarin and other anticoagulants (label): Indomethacin and warfarin act together on bleeding, so the combination carries a higher risk of serious bleeding than either drug alone. (Source 12)
• SSRI and SNRI antidepressants (label): Serotonin released by platelets helps blood clot, so antidepressants that block serotonin reuptake can add to an NSAID's bleeding risk. The label attributes this to case-control and cohort studies. (Source 12)
• Lithium (pharmacokinetic study): NSAIDs raise lithium levels in the blood by reducing how fast the kidney clears it, which can tip someone into lithium toxicity. (Source 13)
• Methotrexate (label): Taking an NSAID with methotrexate may increase methotrexate toxicity, including low white cells, low platelets and kidney dysfunction. (Source 13)

Stopping it

• The label's own guidance is to step the dose back down once the acute phase is controlled, repeatedly, until the person is on the smallest effective dose or off the drug. This is dose minimisation rather than a withdrawal taper: no dependence or withdrawal syndrome is described for indomethacin in the label or in the literature we read. (Source 14)
• For an acute gout attack the label describes a deliberately short course that is stopped abruptly rather than tapered, with the dose reduced rapidly to complete cessation once pain is tolerable. (Source 15)
• Kidney effects of NSAIDs are usually reversible: the label states that stopping NSAID therapy is usually followed by recovery to the pre-treatment state. (Source 16)

What goes wrong

In pooled community observational studies, indomethacin had one of the highest cardiovascular risks of any NSAID studied; because the data are observational this is an association, not a demonstrated cause. (Source 17)

• Systematic review, Low certainty.
• Size: 30 case-control studies with 184,946 cardiovascular events and 21 cohort studies in more than 2.7 million exposed people.
• Who: community NSAID users at typical doses.
• How long: varies by study; risk rose early in treatment.
• Result: indomethacin relative risk 1.30 (95% CI 1.19 to 1.41), compared with naproxen 1.09 (1.02 to 1.16) and ibuprofen 1.18 (1.11 to 1.25)
• Funding: not stated.

Limit of this finding: These are observational studies, not trials, so people prescribed indomethacin may have differed from people prescribed other NSAIDs in ways that also affect heart risk. The randomised-trial meta-analysis that could have settled it did not report indomethacin separately.

Of the less studied drugs etoricoxib, 2.05 (1.45, 2.88), etodolac, 1.55 (1.28, 1.87), and indomethacin, 1.30 (1.19, 1.41), had the highest risks.

In pooled observational studies of upper gastrointestinal complications, indometacin sat in the higher band of NSAID risk; this is an association from observational data, not a trial result. (Source 18)

• Meta-analysis, Low certainty.
• Size: 28 studies met the meta-analysis inclusion criteria, out of 2984 articles identified.
• Who: users of individual NSAIDs compared with non-use.
• How long: varies by study.
• Result: indometacin pooled relative risk 4.14 (95% CI 2.91 to 5.90) for upper gastrointestinal complications, against ibuprofen 1.84 (1.54 to 2.20) and celecoxib 1.45 (1.17 to 1.81)
• Funding: not stated (conducted within the European Community's Seventh Framework Programme SOS project)

Limit of this finding: Observational data again. The pooled relative risk compares people taking the drug with people taking none, so differences between those groups other than the drug could contribute.

4-5 for tenoxicam (RR 4.10; 95% CI 2.16, 7.79), naproxen (RR 4.10; 95% CI 3.22, 5.23), indometacin (RR 4.14; 95% CI 2.91, 5.90) and diflunisal (RR 4.37; 95% CI 1.07, 17.81)

In the largest randomised-trial meta-analysis of NSAIDs, heart failure risk was roughly doubled by all NSAIDs and every regimen increased upper gastrointestinal complications, but indomethacin was not one of the regimens reported, so none of these figures is an indomethacin figure. (Source 19)

• Meta-analysis, High certainty.
• Size: all randomised placebo-controlled and head-to-head NSAID trials pooled in this collaboration; trial and participant counts are in the companion reference cnt-nsaid-2013-methods.
• Who: participants in randomised trials of NSAIDs versus placebo or another NSAID.
• How long: the follow-up of the included randomised trials.
• Result: upper gastrointestinal complications rate ratios: coxibs 1.81 (1.17-2.81), diclofenac 1.89 (1.16-3.09), ibuprofen 3.97 (2.22-7.10), naproxen 4.22 (2.71-6.56); heart failure roughly doubled by all NSAIDs; three more major vascular events per 1000 patients per year on a coxib or diclofenac, one of them fatal.
• Funding: UK Medical Research Council and British Heart Foundation.

Limit of this finding: These are the figures for coxibs, diclofenac, ibuprofen and naproxen. The collaboration did not report indomethacin as a separate regimen, so the only drug-specific estimates for indomethacin in this write-up (cardiovascular 1.30, gastrointestinal 4.14) come from pooled observational studies, which can show an association but cannot establish that the drug caused it.

Heart failure risk was roughly doubled by all NSAIDs. All NSAID regimens increased upper gastrointestinal complications (coxibs 1·81, 1·17–2·81, p=0·0070; diclofenac 1·89, 1·16–3·09, p=0·0106; ibuprofen 3·97, 2·22–7·10, p<0·0001; and naproxen 4·22, 2·71–6·56, p<0·0001).

The US label states that NSAID-caused upper gastrointestinal ulcers, gross bleeding or perforation occurred in about 1% of patients treated for 3 to 6 months and 2% to 4% treated for a year, and that only one in five such events is preceded by symptoms. (Source 10)

• Official position, Certainty not rated.
• Size: not stated in the label passage.
• Who: NSAID-treated patients, with higher risk in the elderly and in those with prior ulcer or bleeding.
• How long: 3 to 6 months and one year.
• Result: about 1% at 3 to 6 months and about 2% to 4% at one year; more than 10-fold increased risk in those with prior peptic ulcer disease or GI bleeding.
• Funding: not applicable (regulatory label)

Upper GI ulcers, gross bleeding, or perforation caused by NSAIDs occurred in approximately 1% of patients treated for 3 to 6 months, and in about 2% to 4% of patients treated for one year.

The label's pooled trial data put headache at 11.7% and mark dizziness, nausea and dyspepsia as occurring in 3% to 9% of patients; everything else in the table is listed as occurring in under 1% of patients, and there is no placebo comparison. (Source 20)

• Official position, Certainty not rated.
• Size: the pooled published controlled trials behind the label's adverse-reaction table; the trial and patient counts are in the companion reference dailymed-indomethacin-ae.
• Who: patients given indomethacin capsules in published controlled trials.
• How long: not stated.
• Result: headache 11.7% and the starred reactions (including dizziness, nausea and dyspepsia) 3% to 9%, both in the table's 'Incidence greater than 1%' column; every other reaction listed sits in the 'Incidence less than 1%' column, and where that first column reads 'None' no reaction in that body system reached 1%.
• Funding: not applicable (regulatory label)

Limit of this finding: The label's table has two columns, 'Incidence greater than 1%' and 'Incidence less than 1%'. Only headache, at 11.7%, and the starred reactions, at 3% to 9%, are in the first column. Every other reaction named under a body system is in the under-1% column, and where the first column says 'None' it means nothing in that body system reached 1%. The long lists of serious-sounding reactions are rare reactions, not common ones. There is also no placebo group, so none of these rates can be compared with what people not taking the drug report.

*Reactions occurring in 3% to 9% of patients treated with indomethacin capsules. (Those reactions occurring in less than 3% of the patients are unmarked.)

In a gastroscopic study the label cites, indomethacin capsules produced significantly more gastric mucosal abnormalities than suppositories or placebo in healthy volunteers. (Source 21)

• Official position, Certainty not rated.
• Size: 45 healthy subjects in the gastroscopic study; 175 patients in the comparative rheumatoid arthritis study.
• Who: healthy volunteers and patients with rheumatoid arthritis.
• How long: not stated.
• Result: significantly higher number of gastric mucosal abnormalities with capsules than with suppositories or placebo; upper gastrointestinal adverse effects comparable between capsules and suppositories in the 175-patient trial.
• Funding: not applicable (regulatory label)

In a gastroscopic study in 45 healthy subjects, the number of gastric mucosal abnormalities was significantly higher in the group receiving indomethacin capsules than in the group taking indomethacin suppositories or placebo.

In a large case-control study of aspirin and ibuprofen rather than indomethacin, heavier drinking raised the risk of major upper gastrointestinal bleeding, and aspirin raised it further at every level of drinking. (Source 22)

• Case-control study, Low certainty.
• Size: 1224 patients hospitalized with acute major UGIB compared to 2945 neighbor controls.
• Who: adults in the United States and Sweden hospitalised with acute major upper gastrointestinal bleeding from new peptic ulcer or gastritis.
• How long: not applicable (case-control)
• Result: relative risk of bleeding rose with alcohol intake to 2.8 at 21 or more drinks per week; among current drinkers, regular aspirin above 325 mg carried a relative risk of 7.0, lower-dose regular use 2.8 and occasional use 2.4, and all of these estimates were statistically significant.
• Funding: not stated.

Limit of this finding: This study measured aspirin and ibuprofen, not indomethacin, and the dose-response it found was for aspirin. The authors state that the ibuprofen estimates did not vary consistently with how much people drank, so the aspirin pattern should not be transferred to other NSAIDs. It is also a case-control study, which shows association rather than cause.

Compared with those who drank less than one drink/wk, the relative risk of acute UGIB increased with increasing alcohol consumption, rising to 2.8 among those who drank > or = 21 drinks/wk.

The label reports that NSAIDs including indomethacin can raise serum potassium even in people with normal kidney function, and that long-term NSAID use has caused renal papillary necrosis and other kidney injury. (Source 23)

• Official position, Certainty not rated.
• Size: not stated in the label passage.
• Who: NSAID-treated patients, with greatest risk in those with impaired renal function, dehydration, hypovolaemia, heart failure, liver dysfunction, diuretic or ACE inhibitor use, and the elderly.
• How long: long-term administration.
• Result: hyperkalaemia reported without renal impairment, attributed to a hyporeninaemic-hypoaldosteronism state; renal papillary necrosis and other renal injury with long-term use.
• Funding: not applicable (regulatory label)

Increases in serum potassium concentration, including hyperkalemia, have been reported with use of NSAIDs, even in some patients without renal impairment.

In the same gout review, moderate-certainty evidence from six trials found that non-selective NSAIDs, the group indomethacin belongs to, probably increase withdrawals due to adverse events and total adverse events compared with selective COX-2 inhibitors. (Source 4)

• Systematic review, Moderate certainty.
• Size: 1,244 participants across 6 trials.
• Who: adults with an acute gout attack.
• How long: not stated in the abstract we read.
• Result: withdrawals due to adverse events RR 2.3 (95% CI 1.3 to 4.1); total adverse events, mainly gastrointestinal, RR 1.9 (95% CI 1.4 to 2.8)
• Funding: not stated in the abstract we read.

Limit of this finding: This compares non-selective NSAIDs as a group against COX-2-selective drugs; the review reports no separate result for indomethacin, so the numbers describe the class and not this drug. The comparison is also drug against drug rather than drug against placebo, so it says which of the two types produced more adverse events, not how often adverse events occur on either.

Non-selective NSAIDs probably increase withdrawals due to adverse events (RR 2.3, 95% CI 1.3 to 4.1) and total adverse events (mainly gastrointestinal) (RR 1.9, 95% CI 1.4 to 2.8).

What the evidence supports

In axial spondyloarthritis, NSAIDs as a class gave clearly better pain relief than placebo at six weeks, with an absolute number needed to treat of 4; the review pools NSAIDs and reports no separate result for indomethacin. (Source 3)

• Systematic review, High certainty.
• Size: 850 participants across four trials.
• Who: adults with axial spondyloarthritis; no trial in the meta-analyses included non-radiographic axial spondyloarthritis.
• How long: six weeks.
• Result: pain in the control group ranged from 57 to 64 on a 100 mm visual analogue scale and was 16.5 points lower in the NSAID group (95% CI -20.8 to -12.2), NNT 4 (3 to 6)
• Funding: not stated.

Limit of this finding: This is class-level evidence. The review pooled different NSAIDs together, so the size of the benefit shown here is not a measurement of indomethacin, and no trial in these meta-analyses included people with non-radiographic axial spondyloarthritis.

High quality evidence (four trials, N=850) indicates better pain relief with NSAIDs (pain in control group ranged from 57 to 64 on a 100mm visual analogue scale (VAS) and was 16.5 points lower in the NSAID group (95% confidence interval (CI) -20.8 to -12.2), lower scores indicate less pain, NNT 4 (3 to 6))

The placebo-controlled evidence for NSAIDs as a class in acute gout comes from a single 30-person trial rated low certainty; the review reports no result for indomethacin specifically. (Source 4)

• Systematic review, Low certainty.
• Size: 30 participants in one trial, within a review of 28 trials and 3406 participants.
• Who: adults with an acute gout attack.
• How long: 24 hours to four days.
• Result: 50% pain reduction at 24 hours in 11/15 on NSAID versus 4/15 on placebo (RR 2.7, 95% CI 1.1 to 6.7), absolute improvement 47% (3.5% more to 152.5% more)
• Funding: not stated.

Limit of this finding: This is class-level evidence. The review pools all non-steroidal anti-inflammatory drugs together and reports no separate result for indomethacin, and the whole placebo comparison rests on a single trial of 30 people that the authors rated low certainty. Do not read it as a measurement of what indomethacin does in gout.

More participants (11/15) may have a 50% reduction in pain at 24 hours with NSAIDs than with placebo (4/15) (risk ratio (RR) 2.7, 95% confidence interval (CI) 1.1 to 6.7), with absolute improvement of 47% (3.5% more to 152.5% more).

Prophylactic indomethacin markedly reduces patent ductus arteriosus in preterm infants in pooled randomised trials. (Source 6)

• Meta-analysis, Moderate certainty.
• Size: the pooled randomised-trial analysis of patent ductus arteriosus in a review of randomised trials and cohort studies.
• Who: preterm infants.
• How long: neonatal period.
• Result: patent ductus arteriosus OR 0.31 (95% CI 0.25-0.38), P < 0.001, I-squared 10% in the randomised-trial analysis.
• Funding: not stated.

Meta-analysis of RCT studies shows infants given prophylactic doses of indomethacin have significantly lower rates of PDA compared to those who did not (OR = 0.31; 95% CI = 0.25–0.38; P -value < 0.001)

In indomethacin-responsive headaches the evidence is an uncontrolled case series: eleven of eighteen patients given an indomethacin trial responded, and response took days to weeks. (Source 8)

• Case series, Very low certainty.
• Size: eighteen patients, eleven of whom responded.
• Who: patients with suspected paroxysmal hemicrania or hemicrania continua at a tertiary headache centre.
• How long: median 13 days to response in hemicrania continua (range 3-35 days); 11 days in paroxysmal hemicrania (range 4-23 days)
• Result: eleven of eighteen (61%) achieved a positive response; most responded to 150 mg/day, with escalation to 225 mg/day or prolonged treatment exceeding four weeks occasionally required; seven patients had negative trials leading to alternative diagnoses.
• Funding: not stated.

The median time to a clinically meaningful response was 13 days for hemicrania continua (range 3–35 days) and 11 days for paroxysmal hemicrania (range 4–23 days). Most patients responded to 150 mg/day, although escalation to 225 mg/day or prolonged treatment exceeding four weeks was occasionally required.

What the evidence does not support

In the same gout review, NSAIDs as a class showed little to no effect on joint swelling after four days and little to no difference in function at 24 hours. (Source 4)

• Systematic review, Low certainty.
• Size: 30 participants in one trial.
• Who: adults with an acute gout attack.
• How long: four days.
• Result: swelling 13/15 on NSAID versus 12/15 on placebo (RR 1.1, 95% CI 0.8 to 1.5), absolute improvement 6.4% (16.8% fewer to 39.2% more)
• Funding: not stated.

Limit of this finding: This is class-level evidence. The review pools all non-steroidal anti-inflammatory drugs together and reports no separate result for indomethacin, and the whole placebo comparison rests on a single trial of 30 people that the authors rated low certainty. Do not read it as a measurement of what indomethacin does in gout.

NSAIDs may have little to no effect on inflammation (swelling) after four days (13/15 participants taking NSAIDs versus 12/15 participants taking placebo; RR 1.1, 95% CI 0.8 to 1.5)

The largest randomised trial of indomethacin prophylaxis found no improvement in death or neurosensory impairment, despite closing the ductus and reducing severe brain haemorrhage. (Source 7)

• Randomized trial, High certainty.
• Size: 574 infants assigned to indomethacin and 569 to placebo with primary-outcome data.
• Who: extremely-low-birth-weight preterm infants.
• How long: to the trial's primary composite outcome assessment.
• Result: 271/574 (47%) died or survived with impairments on indomethacin versus 261/569 (46%) on placebo, odds ratio 1.1 (95% CI 0.8 to 1.4), P=0.61; patent ductus arteriosus 24% versus 50%, odds ratio 0.3, P<0.001; severe periventricular and intraventricular haemorrhage 9% versus 13%, odds ratio 0.6, P=0.02.
• Funding: not stated.

271 (47 percent) died or survived with impairments, as compared with 261 of the 569 infants (46 percent) assigned to placebo (odds ratio, 1.1; 95 percent confidence interval, 0.8 to 1.4; P=0.61)

Across 23 studies, prophylactic indomethacin showed no significant effect on bronchopulmonary dysplasia, pulmonary haemorrhage, intraventricular haemorrhage, necrotising enterocolitis, intestinal perforation, mortality, or length of hospital stay. (Source 24)

• Systematic review, Low certainty.
• Size: 23 randomised trials and cohort studies.
• Who: preterm infants.
• How long: to hospital discharge.
• Result: no significant difference for bronchopulmonary dysplasia, pulmonary haemorrhage, intraventricular haemorrhage, necrotising enterocolitis, intestinal perforation, mortality or length of stay; the review reports high heterogeneity in several of these analyses.
• Funding: not stated.

while no significance was recorded with BPD, pulmonary hemorrhage, intraventricular hemorrhage, necrotizing enterocolitis, intestinal perforation, mortality, and length of hospital stay.

Used as a tocolytic to delay preterm birth, COX inhibitors (mostly indomethacin) showed no clear benefit over placebo or other tocolytics. (Source 9)

• Systematic review, Low certainty.
• Size: 20 studies including 1509 women; indomethacin used in 15 studies.
• Who: women in preterm labour.
• How long: to birth and the neonatal period.
• Result: birth less than 48 hours after trial entry average RR 0.20 (95% CI 0.03 to 1.28; two studies with 70 women) versus placebo; no difference in measures of neonatal morbidity or neonatal mortality; the one small trial showing reduced preterm birth had 36 women (RR 0.21, 95% CI 0.07 to 0.62)
• Funding: not stated.

No difference was shown in birth less than 48 hours after trial entry (average RR 0.20, 95% CI 0.03 to 1.28; two studies with 70 women).

The review's own conclusion is that there is insufficient evidence to decide the role of COX inhibition in preterm labour, because of small numbers, minimal safety data and no long-term outcomes. (Source 25)

• Systematic review, Low certainty.
• Size: the studies pooled in this Cochrane review.
• Who: women in preterm labour.
• How long: no longer-term infant outcomes were available.
• Result: no clear benefit over placebo or any other tocolytic; benefit limited to postponement of birth versus placebo and betamimetics and fewer maternal adverse effects than betamimetics and magnesium sulphate.
• Funding: not stated.

In this review, no clear benefit for COX inhibitors was shown over placebo or any other tocolytic agents.

In the same single placebo-controlled gout trial, adverse events were no different between NSAID and placebo, and the interval around the estimate spans both fewer and more adverse events, so no difference was shown either way. (Source 4)

• Systematic review, Low certainty.
• Size: 30 participants in one trial, within a review of 28 trials and 3406 participants.
• Who: adults with an acute gout attack.
• How long: 24 hours to four days.
• Result: no withdrawals for adverse events in either group; two adverse events (nausea and polyuria) in the placebo group, RR 0.2 (95% CI 0.0, 3.8), with absolute difference of 10.7% more (13.2% fewer to 38% more)
• Funding: not stated in the abstract we read.

Limit of this finding: The confidence interval runs from 13.2% fewer adverse events to 38% more, so it includes no difference at all. With two events among 30 people the trial could not detect a difference in harm in either direction; this is not evidence that NSAIDs are safe in gout, and it is class-level, not indomethacin-specific.

NSAIDs may result in little to no difference in withdrawals due to adverse events (0 events in both groups) or in total adverse events; two adverse events (nausea and polyuria) were reported in the placebo group (RR 0.2, 95% CI 0.0, 3.8), with absolute difference of 10.7% more (13.2% fewer to 38% more).

In the same case-control study the ibuprofen estimates did not vary consistently with how much people drank, so the aspirin dose-response was not reproduced for the other NSAID it measured. (Source 22)

• Case-control study, Low certainty.
• Size: 1224 patients hospitalized with acute major UGIB compared to 2945 neighbor controls.
• Who: adults in the United States and Sweden hospitalised with acute major upper gastrointestinal bleeding from new peptic ulcer or gastritis.
• How long: not applicable (case-control)
• Result: regular ibuprofen use at all doses combined gave a significantly elevated estimate of 2.7 among all drinkers combined, while occasional ibuprofen use was not associated with bleeding (1.2); the authors describe the ibuprofen data as more limited.
• Funding: not stated.

Limit of this finding: Neither drug here is indomethacin, and the authors say there were insufficient data to evaluate other NSAIDs by alcohol consumption at all. So the honest position for indomethacin plus alcohol is the label statement, not a measured risk.

Data for ibuprofen were more limited, but the relative risk estimates did not appear to vary consistently with level of alcohol consumption. For regular use (all doses combined), the estimate among all drinkers combined was significantly elevated, at 2.7; occasional ibuprofen use was not associated with UGIB (1.2).

Where the evidence is mixed

In the same class-level review, withdrawals for adverse events and serious adverse events did not differ significantly between NSAIDs and placebo over 12 weeks, which is a short observation window for NSAID harms. (Source 3)

• Systematic review, High certainty.
• Size: 1165 participants across five trials (withdrawals) and 671 across three trials (serious adverse events)
• Who: adults with axial spondyloarthritis.
• How long: 12 weeks.
• Result: withdrawals due to adverse events 39/1000 on NSAID versus 52/1000 on placebo (RR 0.75, 95% CI 0.46 to 1.21); serious adverse events 3/1000 versus 2/1000 (RR 1.69, 95% CI 0.36 to 7.97)
• Funding: not stated.

Limit of this finding: Class-level again: NSAIDs are pooled and indomethacin is not reported separately. Twelve weeks is also far too short to see the cardiovascular and gastrointestinal harms that the boxed warning is about, so this is not evidence of safety.

indicates that withdrawals due to AEs and number of serious AEs did not differ significantly between placebo (52/1000 and 2/1000) and NSAID (39/1000 and 3/1000) groups after 12 weeks

Where the research disagrees

Whether indomethacin should still be in routine clinical use at all, given its comparative safety profile

• McGettigan and Henry, systematic review of community observational studies (2011), systematic review of 30 case-control and 21 cohort studies: Indomethacin is an older, rather toxic drug, and the evidence on cardiovascular risk casts doubt on its continued clinical use. (Source 26)
• US prescribing information (2026 label), regulatory position in the 2026 US prescribing information, which retains five approved indications: After the acute phase of the disease is under control, an attempt to reduce the daily dose should be made repeatedly until the patient is receiving the smallest effective dose or the drug is discontinued. (Source 14)

Whether prophylactic indomethacin in extremely preterm infants is worth giving

• TIPP randomised trial (Schmidt and colleagues, 2001), randomised controlled trial, 1,202 infants, 18-month outcome: Indomethacin reduced the incidence of patent ductus arteriosus (24 percent vs. 50 percent in the placebo group; odds ratio, 0.3; P<0.001) and of severe periventricular and intraventricular hemorrhage (9 percent vs. 13 percent in the placebo group; odds ratio, 0.6; P=0.02) (Source 7)
• Al-matary and colleagues, systematic review and meta-analysis (2022), systematic review and meta-analysis of 23 randomised trials and cohort studies: Since the meta-analysis results regarding effectiveness of prophylactic indomethacin varied based on the study design particularly with regard to outcomes such as surgical PDA ligation and severe IVH, this warrants the need for more evidence regarding the effectiveness of prophylactic indomethacin in very low birth weight infants. (Source 24)

How much

• Reference intake: There is no reference intake for a drug: dosing is set by the prescriber. As a position, the 2026 US label gives indomethacin capsules 25 mg twice or three times a day for rheumatoid arthritis, ankylosing spondylitis and osteoarthritis, increased in 25 mg or 50 mg steps at weekly intervals if symptoms continue. (Source 5)
• Upper limit: As a position, the 2026 US label describes increasing only until a total daily dose of 150 mg to 200 mg is reached, and states that doses above this generally do not increase the drug's effectiveness. (Source 5)
• Studied: The TIPP trial gave 1202 infants of 500 to 999 g birth weight indomethacin 0.1 mg per kilogram of body weight intravenously once daily for three days. (Source 27)
• Studied: In the tertiary-centre headache series, most responders were treated at 150 mg/day, with escalation to 225 mg/day in some. (Source 8)
• Studied: The label's acute gout regimen, as a position, is 50 mg three times a day until pain is tolerable, then rapid reduction to complete cessation. (Source 15)

A common belief, and what the research shows

The belief: Indomethacin is just a strong ibuprofen, so if ibuprofen is safe enough then indomethacin is too.

What the research shows: The comparative literature separates NSAIDs. In pooled community observational studies, ibuprofen carried a cardiovascular relative risk of 1.18 (1.11 to 1.25) while indomethacin was 1.30 (1.19 to 1.41), and the review's authors wrote that "Indomethacin is an older, rather toxic drug, and the evidence on cardiovascular risk casts doubt on its continued clinical use." For upper gastrointestinal complications, pooled observational data put ibuprofen at 1.84 (1.54 to 2.20) and indometacin at 4.14 (2.91 to 5.90). The class boxed warning applies to both, but the size of the risk is not the same.

Questions and answers

What is it?

Indomethacin is a prescription non-steroidal anti-inflammatory drug. It has painkilling, anti-inflammatory and fever-reducing effects. It comes as capsules, an oral suspension, suppositories and an intravenous form used in newborn babies. It is one of the older NSAIDs and is still used for inflammatory arthritis, gout and, in hospital, to close a blood vessel that should have shut after birth. (Source 1)

What does it do in the body?

It blocks the two cyclooxygenase enzymes, COX-1 and COX-2, so the body makes fewer prostaglandins. Prostaglandins sensitise nerve endings to pain and drive inflammation, so less of them means less pain and swelling. The same prostaglandins protect the stomach lining and help keep blood flowing through the kidneys and, in a newborn, keep the ductus arteriosus open, which explains both the drug's uses and its harms. (Source 1)

Is it good or bad for you?

Both, and it depends on the setting and the length of use. For a short course in an inflammatory joint condition or a gout attack, trials show real symptom relief. For longer use, it carries a boxed warning for heart attack, stroke and gastrointestinal bleeding, and observational meta-analyses place it among the higher-risk NSAIDs for both. In preterm babies it closes the ductus arteriosus reliably but did not improve survival free of impairment in the largest trial. (Source 2)

How do you get more of it?

There is no way to get more of it other than a prescription: it is a manufactured drug, not a nutrient, and nothing in food contains it. Dose and frequency are decided by the prescriber, who raises the dose in steps if symptoms continue. Taking more than prescribed does not add effectiveness; the label states doses above the stated range generally do not increase the drug's effectiveness. (Source 5)

If it is harmful, what reduces it?

If indomethacin is causing harm, the drug is stopped or the dose reduced, under medical advice. Many of its effects reverse: the label states that stopping NSAID therapy is usually followed by recovery to the pre-treatment state for the kidney effects. Serious adverse reactions are a reason to stop the drug outright rather than reduce it. Nothing is taken to clear it faster. (Source 16)

Why might someone be low in it or missing it?

Does not apply in the sense this question is usually asked. Indomethacin is not something the body makes or absorbs from food, so nobody is naturally low in it. A person has it in their system only if it has been prescribed and taken, and prescribers deliberately aim for the smallest dose that works, so a low dose is often intentional rather than a deficiency. (Source 14)

Which whole foods contain it or feed it?

No whole food contains indomethacin or feeds it. It is a synthetic drug. We searched the prescribing information and the trial and review literature for this group and found no food source and no quantified food interaction for indomethacin. (Source 1)

We searched: the 2026 DailyMed full prescribing information for indomethacin capsules, including the clinical pharmacology and drug interactions sections, and Europe PMC for indomethacin food-effect studies

What happens if you do not have it?

Nothing happens from not having the drug itself; what happens is that the condition it was treating goes untreated. In the Cochrane axial spondyloarthritis trials, people on placebo sat at 57 to 64 on a 100 mm pain scale while those on an NSAID were about 16.5 points lower, so without it pain and stiffness are typically worse. In a preterm baby with a patent ductus arteriosus, half of untreated infants in the TIPP placebo group still had the duct open. (Source 3)

How can you test for it?

There is no routine blood test for indomethacin levels and none is used to guide dosing. What is monitored instead is harm: the label says to check haemoglobin or haematocrit if there are signs of anaemia, since bleeding into the gut may be silent, and to monitor kidney function in people at risk. Only about one in five people who develop a serious upper gastrointestinal event on an NSAID have warning symptoms first. (Source 28)

References

1. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
2. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
3. Cochrane Database of Systematic Reviews. Non-steroidal anti-inflammatory drugs (NSAIDs) for axial spondyloarthritis (ankylosing spondylitis and non-radiographic axial spondyloarthritis). 2015. PMID 26186173, DOI 10.1002/14651858.cd010952.pub2. Read the source
4. Cochrane Database of Systematic Reviews. Non-steroidal anti-inflammatory drugs for acute gout — abstract, Main results. 2021. PMID 34882311, DOI 10.1002/14651858.cd010120.pub3. Read the source
5. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
6. Frontiers in Pediatrics. Use of Prophylactic Indomethacin in Preterm Infants: A Systematic Review and Meta-Analysis. 2022. PMID 35463887, DOI 10.3389/fped.2022.760029. Read the source
7. New England Journal of Medicine. Long-term effects of indomethacin prophylaxis in extremely-low-birth-weight infants. 2001. PMID 11430325, DOI 10.1056/nejm200106283442602. Read the source
8. Neurological Sciences. Indomethacin testing in suspected paroxysmal hemicrania and hemicrania continua: Clinical phenotypes and response patterns in a tertiary headache centre. 2026. PMID 42834248, DOI 10.1007/s10072-026-09456-9. Read the source
9. Cochrane Database of Systematic Reviews. Cyclo-oxygenase (COX) inhibitors for treating preterm labour. 2015. PMID 26042617, DOI 10.1002/14651858.cd001992.pub3. Read the source
10. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
11. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
12. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
13. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
14. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
15. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
16. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
17. PLoS Medicine. Cardiovascular risk with non-steroidal anti-inflammatory drugs: systematic review of population-based controlled observational studies. 2011. PMID 21980265, DOI 10.1371/journal.pmed.1001098. Read the source
18. Drug Safety. Individual NSAIDs and upper gastrointestinal complications: a systematic review and meta-analysis of observational studies (the SOS project). 2012. PMID 23137151, DOI 10.2165/11633470-000000000-00000. Read the source
19. The Lancet (Coxib and traditional NSAID Trialists' (CNT) Collaboration). Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. 2013. PMID 23726390, DOI 10.1016/S0140-6736(13)60900-9. Read the source
20. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label) — 6.1 Clinical Trials Experience, Table 1 (LOINC 42229-5). 2026. Read the source
21. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
22. American Journal of Gastroenterology. The risk of acute major upper gastrointestinal bleeding among users of aspirin and ibuprofen at various levels of alcohol consumption — abstract, Methods and Results. 1999. PMID 10566713, DOI 10.1111/j.1572-0241.1999.01517.x. Read the source
23. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source
24. Frontiers in Pediatrics. Use of Prophylactic Indomethacin in Preterm Infants: A Systematic Review and Meta-Analysis. 2022. PMID 35463887, DOI 10.3389/fped.2022.760029. Read the source
25. Cochrane Database of Systematic Reviews. Cyclo-oxygenase (COX) inhibitors for treating preterm labour. 2015. PMID 26042617, DOI 10.1002/14651858.cd001992.pub3. Read the source
26. PLoS Medicine. Cardiovascular risk with non-steroidal anti-inflammatory drugs: systematic review of population-based controlled observational studies. 2011. PMID 21980265, DOI 10.1371/journal.pmed.1001098. Read the source
27. New England Journal of Medicine. Long-term effects of indomethacin prophylaxis in extremely-low-birth-weight infants. 2001. PMID 11430325, DOI 10.1056/nejm200106283442602. Read the source
28. DailyMed, U.S. National Library of Medicine (labeller: A-S Medication Solutions; manufactured by Hetero Labs Limited). INDOMETHACIN (indomethacin) capsule - FDA prescribing information (full label). 2026. Read the source